一种基于在线大数据的高频CPI指数的设计及应用

Design and Application of Novel CPI Based on Online Big Data

在数字经济时代探索如何运用在线大数据编制实时高频物价指标。研究方法:设计了中国第一套基于互联网在线大数据的居民消费价格指数,从多方面分析指数质量及其应用。研究发现:在线iCPI可实现各层次类别的日、周、月指数无滞后实时更新;指数数据从采集、清洗到加工和发布均由计算机自动进行,既节省人力又减少人为干预因素;指数在代表一般物价变化、精准捕捉典型事件影响、现时预测通货膨胀、实时监测宏观经济形势等方面表现突出。研究创新:首次采用在线大数据编制CPI,弥补了中国尚无实时高频物价指标的空白。研究价值:在线iCPI是传统CPI的有益补充,其编制思维和方法可用到更多高频宏观经济指标的构建上。

Research Objectives: The paper aims to explore the design and application of real-time high-frequency price indicators with online big data in the era of digital economyResearch Methods: We design the first set of Internet-based Consumer Price Indices (iCPI) in China, and analyze the index quality and its applications from multiple aspectsResearch Findings: Firstly, iCPI can realize real-time updating of daily, weekly, and monthly indices of various levels of products and service (see official website www.bdecon.com)Secondly, iCPI is automatically generated in the computation procedure from the data collection, cleaning, processing and final online publishing, which effectively saves time and avoids human interventionThirdly, high-frequency iCPI performs well in representing general price changes, accurately capturing the effects of special events, nowcasting the inflation and real-time monitoring the macroeconomic situationResearch Innovations: We firstly employ online big data to design CPI in China, which makes up for the research gap of real-time high-frequency price indicatorsResearch Value: Online iCPI is the beneficial complement of official CPI, and the proposed realization roadmap can be applied to construct different high-frequency macroeconomic indicators in the future.